Stockfish Has Ruined Chess
Vložit
- čas přidán 16. 12. 2022
- Puzzle FEN:
6bk/7p/2q1PP2/2Pp4/3R4/5K2/8/B7 w - - 0 1
Puzzle Details:
Evreinov, 1959, First Prize
Get my 5 most popular courses in one discounted bundle⏩chessvibescourses.thinkific.c...
☑️ Play Chess Here:
www.chess.com/membership?ref_...
Links are affiliate links and help support the Chess Vibes channel via a commission. - Jak na to + styl
Stockfish finding an even more complex solution to an already beautiful problem is hardly "ruining the game"
it's not a solution though, since it ends in a draw. But yes I agree, it's not ruining.
@@BigDBrian black was losing, but ended up getting a draw, so it's a solution, just not for white
@@whackygamer19 Yes
But, when you look at it logically, attacking bishop straight away is not at all that complex move, no?
People keep saying there's a black pawn on a4.
Back over 60 years ago, even if a solution took 2 years to process, a faster computer released 1 year later could actually find the solution first. The transition to transistors changed everything.
@@hrsmp and knew that the transition will happen
@@aggressivepvp
At least 55 years ago, when I was in college, everybody who was at all knowledgeable about computers knew that the technology was progressing so fast that next year's computer would render this year's obsolete. A teacher told us, tongue-in-cheek, that if you ordered a computer today (back then) and another in a month, the second computer would be cheaper, more powerful, and would arrive sooner. A bit of hyperbole, but illustrating the fast pace of development. This, of course, was long before microchips. Yeah, we knew it was going by leaps and bounds.
@@danielschechter8130 Thank you for sharing your experience with us :) I was born in 94
It's very important to bear in mind that having faster hardware is only a piece of the solution. The equally big piece is figuring out how to write the software that allows a computer to play chess efficiently. I am a retired programmer who majored in Computer Science in the 1970's and I can tell you that the algorithms needed to do this hadn't been figured out yet.
For example, when I was in school, we students were asked things like, how would you represent a chess position in a computer? How would you evaluate that position? Find threats? Discover attacks? Look several moves ahead? Figuring out newer and better ways to do this and more, repeatedly, has been a large part of the work!
I also want to mention, for anyone reading this who isn't into computers, that hardware and software advances occur together. While transistors were a physics discovery, modern computer hardware is only possible with advanced software. As hardware advances occur it gives us programmers more time and space (i.e. processing power and computer memory), and then we come up with solutions that wouldn't have been feasible previously. So we then write more sophisticated software, which is then used to create even better hardware, and so on, and we also are able to make better chess algorithms. Bottom line: both are needed.
I thought Moore's Law said computing speed doubled every 18 months...
Yep, searched his name and found the puzzle. There is indeed a black pawn on a4.
Yes it is
Yep, Nelson messed up
@@ElSantoDeMostoles "CZcamsrs are ruining chess" 😅
The thing is that even in the original puzzle Stockfish finds a way to defend itself for 11 more moves than in the "solution"
Instead of Bf7 it plays h5
@@ElSantoDeMostoles but in the page show have the line h5 too. Is longer safe but was not the main line of the puzzle
My $5 says that Nelson got the puzzle from 200 Brilliant Endgames By: Irving Chernev, which has the puzzle printed without the a4 pawn 😅
So there should be a black pawn on a4?
@@Dhrazor Yeah, which stops stockfish's plan in this video.
nelson apology video incoming
yeap
Oh, thats silly, so the actual thing that was wrong was that one book (if thats the case, which probably is)
Hey Nelson, Several people in the comments say that Black has a pawn on a4 but is not shown here. That might be the difference. First time seeing this though.
@Samuel Laufer the pawn is black
We are adding pieces to black to get white to win. Really funny 🤣.
@@tapiolouhi9047 The actual puzzle had a Black pawn on a4. Nelson forgot to put the pawn on a4 when showing the puzzle. I have already looked it up.
@@tapiolouhi9047 That's not even rare in puzzles. I believe there is even a genre that is kind of the opposite, where white has to get rid of a white piece (while leaving everything else the same) in some complicated way in order to succeed.
@@tapiolouhi9047 idk ive looked at my own pieces like "sweet Buddha i wish i could stab my own pawn"
There’s a black pawn on a4 in the original puzzle which prevents the queen idea
That would make the original puzzle flawless again...
The pawn would have to be on a3.
No, a white pawn would have to be on a3, a black pawn could be on a4.
@@johnkirby3?? There doesn't have to be any white pawn on the a file
@@64MilestotheGallonyeah, nelson did a lil mistake and didn’t put the pawn in the a file
If you’d like to see the puzzle and how it originally included a pawn on A4, the Arves Chess endgamestudy association has a whole webpage dedicated to the end games of Viktor Alexandrovich Evreinov
You forgot to put the black pawn on A4.
How he ruined (a) chess (study) hahahahah
A white or black pawn anywhere along a2-a4 fixes this puzzle. (Edit: if the pawn is on a4, it has to be black, because a white one can simply be captured).
Yeah, that is a nice, simple fix to this puzzle.
so the youtuber wasted our time?
@@fs.jahahajjako8826 He made a mistake. It happens
@@fs.jahahajjako8826 It is a nice puzzle, and he shows the lines. I don't think it was a waste of time to watch this. The correction suggested above helps make the lines shown work properly, as imagined by the designer.
@@fs.jahahajjako8826 It was a mistake. There was actually a pawn on a4.
@Chess Vibes the video that you made also has a flaw where there wasn't an A4 black pawn.There is a comment which said about this issue and I find out that it is indeed true.
I added the black pawn on a4 that was supposed to be there and stockfish found a way for black to delay it for longer.
there is supposed to be a pawn on a4 buddy
My idea was that instead of moving the bishop to block, black can just move the pawn forward that is next to the king. if you check the king now, it can just move. There's no checkmate.
You would just take the Queen with your own new Queen next move with checkmate, as it's still under attack by your bishop and only defended by blacks King.
Same
Black needs to move the pawn to H5 otherwise it would still be checkmate in 2 by white underpromoting to Knight. This does prevents checkmate for a while, but stockfish still says that it will be eventually checkmate by white since black still allows white to promote to Queen with easy access to checking the opposing King, while black doesn't.
If you try h5 instead of Bf7 for Black, white goes f7+ and when Black moves white plays f8 = Q which looks very dangerous for Black
@@marvinabt4964 What is the queen under an attack from? Where is the threat to c6?
Hello Nelson!
Good news Victor Alexandrovitch didn't make a mistake. In his winning study he put a black pawn on a4 square. Take a look at the picture in Russian Wikipedia related to this person.
So Stockfish still didn't ruin chess that much! :)
This guy forgot to place black pawn on a3💀💀💀
a4
a3, a2 and a4 are all work
More like stockfish is UNREASONABLY good at using knights and queens, honestly. Their potential strings of movements/forks are just too complex for humans to grasp 30 moves ahead.
I like how you can see the rubber duck chess board that he posts every day
You got the puzzle wrong there is a pawn on a4
I haven't seen this problem with a full citation but could it be that it was first published with no bPa4, then someone found that ...Qa4 refutes, then Evreinov added bPa4 and the problem was later published with this correction? It often happens that a problem is published which is a correction of a problem published earlier.
@@rosiefay7283 At very least in the page the CZcamsr show have Bpa4, and explain the line h5
As a complete chess noob, pawn to F7 actually came to my mind first when I saw the puzzle, this is oddly validating lol
Me too, i play on lichess (criticalcess123)
Lol same it seemed pretty obvious and as the video wen ton I accepted it as wrong only for stockfish to come back and say it was the right move.
Im having a hard time seeing whats wrong with the line starting with rg4. Any help would be appreciated :)
The fault lies not in Stockfish, but ourselves. -- Shakespeare, probably
Seems pretty easy to fix this as a puzzle problem. The solution works again as well if there is an extra pawn (from either side, but make it black so it seems more threatening) blocking the Queen's possible attack on the bishop. Or does stockfish find a new alternative?
I wonder what stockfish would think of moving the a4 black pawn forward before moving the queen to a4 (original puzzle had a black pawn on a4 according to someone)
I let Stockfish analyse the game with the black a4 pawn and it is a win for white, but much more complex. Instead of playing Bf7, black plays h5 to create space for the king. From there it's a mate in 17 for white: 2. f7+ Kh7 3. f8=Q Qc7 4. Qf5+ Kh6 5. Qf6+ Kh7 6. Qh8+ Kg6 7. Qxg8+ Kf5 8. Qf7+ Qxf7 9. exf7 Ke6 10. f8=Q Kd5 11. Qd6+ Kc4 12. c6 e2 13. Qe6+ Kd3 14. c7 Kd2 15. c8=Q e1=Q+ 16. Qxe1+ Kxe1 17. Qc2 a3 18. Bc3#. So even with the black a4 pawn, this is quite a bit of an endgame and imo not really an elegant puzzle.
i aint reading allat
I haven't checked this with Stockfish, but can't you get the intended effect if you start the puzzle with a black pawn on a4, or maybe a white pawn on a3?
Yeah that’s what the original puzzle is in this video he used a version that printed the puzzle wrong
At 7:15 rook to G4 can be a checkmate as only move left is pawn to D4 and then white bishop to D4 is just checkmate
I saw this too
the line keeps going from there, because blacks pawn D4 comes with check, and white will never have time to pick up the pawn with the bishop.
I was about to comment the same thing, but then i found the check.
Yeah, doesn't work because Pawn to D4 is check by the black queen, you'll never get a chance to check the opponents king again.
You know, I’ve done so many chess puzzles over the years where I’m scratching my head over flaws like this one. I always thought it was me that was missing something.
It should be a black pawn on a4. Wrong starting position of the study and wrong title of the video. Please correct.
There's a black pawn on a4. I believe you missed it?
One day we will discover the last position you showed wasn't a draw...
I honestly would love to see a video of more chess problems run through stockfish
What about improving this puzzle by, for example, adding black pawns which will prevent the queen from moving Qa4?
In the original there was a black pawn on a4
Looks like Stockfish fixed chess yet again.
My question is knowing that Stockfish found this flaw, is there a way to fix the puzzle so it works the way it’s intended? I know it wouldn’t be the same thing in spirit but could the addition of a pawn at a2 (or something like that) prevent the bishop attack enough to allow the puzzle to run its course?
As others have said, this puzzle was misprinted. There is supposed to be a black pawn in A4, which stops Stockfish's idea. We all make mistakes.
f7 would've been my opening move in that position because moving the rook allows black's pawn to advance, blocking the bishop's diagonal.
I agree. Blacks only move that doesn't get mate in 1 is to move the h7 pawn
It's not ruining. It's revealing. Revealing the flaws and maybe even hints of corruption.
What about:
1.Pf7 - black will attempt to stop the promotion by playing either BxP or Qc8/Qe8
2. Rg4 (discovering check from bishop, black king unable to move, so has to play Pd4)
3. BxP checkmate
If black plays 1.... Qa6 then
2. Rg4 check Qxa1
3. Pxg8 (promote to rook or queen) mate
And if black plays 1........ Qd6 (preventing the bishop from taking the pawn when it moves down), then
2. Rg4 check Pd4
3. Pxg8 (promote to Q or R) mate
at 7:14 how do you stop Rg4? (you can go back a few secs to see what i mean/ a move or two)
1.f7 Bxf7 2.Rg4+?? loses to the discovered check 2...d4+.
Hi Nelson
I think we can save this position by simply placing the pawn on a2 or a3
6bk/7p/2q1PP2/2Pp4/3R4/5K2/P7/B7 w - - 0 1
Stockfish is basically the repository of all seen possibilities and variations, as seen through the minds of every great player whom contributed recordable data...it's like the ability to compose wondrous music without ever sounding a note personally. Stockfish is the repository of the artistry, which exists within us.
Nope, stockfish doesn't use the repository of seen possibilities but basically tries every possibility and checks the success rate
You come to correct a sentiment?
Don't bother
Maybe the analogy was there just for the hell of it
It's like the idiotic axiomatic ponderous statement Morpheus makes about the Matrix....no body can tell you what the Matrix is....
@@burtonaka... you write like a 12 year old kid. and you have no idea of how stockfish actually works
In my own self-analysis of the position I found h5 on move 2 of the puzzle which leads to a whole line of f7+ Kh7, f8=Q e3+, Kg1 Qc7, Qf5+ Kh6, Qf6+ Kh7, Qh8+ and white wins the bishop and presumably will go on to win the game.
This is all assuming there's a Black pawn on a4, as in the original puzzle.
I assume the original puzzle does have this line but it wasn't mentioned in the video and on engine analysis it's the top Stockfish line.
Nah, instead of f8=Q on third move you'd play f8=N++
@@last-chance8476 Aside from the emotional damage you deal from promoting to a Knight with check, this would be a game-losing blunder.
@@isaakvandaalen3899 it's not check, it's check mate going by your previous moves.
@@last-chance8476 Like I said, I checked stockfish and this is the top line. There is no faster checkmate assuming Black defends properly. Perhaps you're referring to a different position.
Chess puzzles uses to be so scuff, stockfish really helps remove bias and nonsense from them and not leave intelligent people scratching their heads for hours.
But if you just add a pawn in front of the bishop, the ouzzle works again, am I right?
Yeah, the biggest problem with great strategies is that for some reason the opponent doesn’t want to follow them 😢
I have a question, can't black move his pawn in front of the king so that the king could escape? Please correct me if I am wrong
I like how stockfish's line that started with the pawn was my first thought seeing the board
I was legit so confused at why Nelson didn't explain Qa4 there cuz that's what I would have played as black, seems like its the flaw lol
i am pretty new to puzzles, and this learned me, that not only the main figurs are important. also the pawns blocking the figures
Something else, I think, depending on the skill of your opponent, is
Pawn:f7, Bishop×f7 to prevent promotion, Pawn×f7, Queen:C8 to prevent promotion, Rook×d5 leads to Checkmate with the bishop. King can't move forward because of Pawn, can't move out of corner because Bishop covers the diagonal and Pawn covers the sidestep.
It's obv not the solution since a better player would see it coming, but it's what I saw when looking at it lol
Am looking at this right? After 1. f7, Bxf7 Don't capture with the pawn, 2. Rg4 check, d4 3. Bxd4 mate.
I missed 2. d4 is check.
the position is incorrect, you forgot to place a black pawn on a4
"That is something that would not have happened back in 1959"
"How did nobody question this move...how was this missed"
Can't rook go get and then push the pawn? Plz someone answer
I agree that it seems like Qa4 is a strange thing to overlook, especially since that simply adding a white or black pawn on a2 or a3 could possibly solve the flaw in the puzzle
As you may have noticed (if not, read the comments), the original puzzle had a black pawn on a4. The reason why Stockfish could win the position is that it's not the position from the puzzle.
Maybe there was supposed to be a pawn in a-line blocking the queen attack, but was forgotten to place there. White pawn in a2 or a3 or black in a4 or any combo of them would do the trick.
Seems like there's conflicting sources on if there's a pawn on a4 or not. Some pictures do show it, some don't.
i have tried it against stockfish and its not working. because after the e3 pawn checks open the diagonal of the queen, king e1, queen a4 threatens the bishop then the bishop goes wherever the queen will take it, because if the bishop goes to b2 or c3, then the queen will check first to d1, then the king will definitely advance to square two, then the queen checks again to c2 and wins the bishop. if the bishop goes to c4 then the queen immediately takes it. if bishop goes to e5 then queen d1 checks, then king h2 or g2, if king h2 then queen h5 checks and take the bishop. but if the king is g2, then the queen is d5 check and take the bishop. besides that, black can also move the h5 pawn so that there is no checkmate.
in this position clearly won by black (correct me if I'm wrong).
A white or black pawn anywhere along a2-a4 fixes this puzzle
Sorry if my English is bad because I'm an Indonesian chess player and I'm using google translate.
5:11
I think the flaw is, if I were playing Black, after White's rook moves to e4, I'd move my pawn down to d4. This blocks the bishop's pin, and creates a queen's pin against White, since they now cannot move the rook to capture the pawn.
Now, bishop CAN take the pawn, which gives it center board while regaining the pin, so from there I think the game might go differently...
We are using the version he is using so we do not have a black pawn at a4. Now then, based on what you're saying I would go Bxd4. Now you're getting mated on next move so your best idea would be Qa4 attacking the bishop. f6 to f7 discovered attack check, you are forced to Qxd4. Rxd4.
Now you're most likely gonna go Bxf7 and so e6xf7. Kg7, Rf4 protecting the pawn that is about to promote. Kg6 or h6 doesn't matter because you will get ladder mated by rook and queen.
In case you decide after Bxd4 to move the bishop to f7 you get taken and your king cannot escape, meaning you have to Qxf6 check, which simply leads to you being taken by the bishop and getting mated.
Unless I missed some other legal move (or made any mistakes, I am currently half asleep staring into my phone) that does anything in these situations I see it as quite hopeless.
Yes the end game would go differently because you've just set yourself to be very quickly check mated after the bishop takes your pawn...
At 7:16 why does stockfish have you take the bishop?
Isn’t rook to g4 checkmate?
Best move for black after pawn f7 would be to pawn h5 instead of taking with the bishop (moving to h6 would set up a mate in 2) or queen a4
Or am I overlooking something?
What's wrong with pawn f7?
Doesn't matter what the opponent is doing, rook to g4 and its discovered check?
Am I overlooking something?
d4 is a discovered check on the white king and blocks the bishop. In the long run, that variation is a win for black.
It occurs to me that after black moves the queen, white should move c6 to free up another diagonal for the bishop. With the white pawn holding the black king on row 8, the white bishop might be able to put the black king in checkmate. I haven't tried it out, so it may not work.
Would it fix the puzzle if we gave white a pawn on b3? Or alternatively maybe give black a pawn on a4? Something that stops the queen move.
I like pawn to f7, black has no check depending on black move, follow rook to g4 if why does not mate, white should promote. What does Stockfish do? Push the pawn?
Question: pawn to f7, IF black takes it with bishop, then rook to g4, checkmate?
"If closer examination breaks something, then it was already broken to begin with."
I do wonder... Would adding a white pawn to b3 make the puzzle work?
My 800 elo brain immediately knew how to solve this puzzle, I also just did not see the checkmate threat and thought "ooga booga attack Bishup" but I'm taking this W.
If the queen attacks the e pawn and checked the rook, just temporarily move the rook to g3 or g2 where the king is protecting it try to checkmate it
What about moving the pawn in front if king instead of BxF7?
one pawn in a2 should fix the problem right?
Also at any point Black moves H7 pawn the turn they figure the bishop would move. and check mate is no longer possible with that move set. So ignoring the queen take bishop there is still a vary clear stall that can happen anyways. I mean they could rush E6 up but black's queen takes it before promote.
I found that trying to figure it out first so I figured I can't find checkmate because of the pawn move anyways. White doesn't have enough pressure to checkmate.
Hey, Nelson. I have a question. Why can't Black move his h pawn?
Thanks for the video and your videos in general! They're always fun to watch and also I like the old puzzles a lot, much more than "modern studies" which are often so chaotic that it's almost impossible to follow. In my opinion, this is really something which was ruined by computers - chess composers are no longer satisfied with one or two mindblowing ideas in one study but they want to put "everything" into one single puzzle so the lines get ridiculous, unmanageable and therefore boring (for me).
I didn't believe that such a simple refutation could have been missed back then, so I checked it and the flaw is actually somewhere else, namely in the starting position because a black pwan on a4 is missing which is actually there in the original study (easy to google). So everything's fine with this great study :)
Nevertheless, there are definitely old puzzles which turned out to be flawed by computers. However, many of these flaws can be fixed by slightly adapting the starting position, so I don't think that Stockfish really "ruined" it :)
If you put a pawn on a3 problem is solved so not a big problem?
im bad at chess, what stopping black from pushing the h pawn and avoiding checkmate?
Solution should still work with a pawn on a3, right?
I was thinking about pawn push a second after seeing position but i didn't knew it would lead to draw
Can we fix this by adding a pawn to B6 or something?
Didn't see qa4, but in that position I was wondering why play the bishop instead of qa8.
Qa4 was the 1st move I looked at when I saw the base puzzle.
But it was protected by the rook so I ignored it.
It is mistake by chess vibe queen can’t go on a4 because in that problem there is black pawn on a4!!! Mistake by you .
Can someone explain why moving pawn to f7 as the first move isnt the answer
I also thought exchanging the pawns would be best play without considering the rook move.
Was the second place puzzel any better?
Pawn to f7 and rook to g4 is a stunning checkmate any corrections please....???
7:38, what is stopping bishop from putting king in check, forcing it to take and allowing pawn to promote?
Throughout the whole video i was trying to figure out why isn't moving the f6 pawn the winning move.
Move pawn to F7. If bishop takes it you can take it with another pawn then queen will either block promotion of pawn or can kill bishop. If queen moves you can kill pawn using rook and it will be a check
What happens if black moves the h pawn?
What if black pushes the h pawn to avoid checkmate?
Weird I spotted attacking the bishop immediately. But there follows a sequence requiring very high accuracy; I think that's what I would mess up.
The move I guessed at the start was f7 right away. I feel very proud of myself lol
The technical term for this kind of thing in the world of puzzle competitions is a "cook". That it was overlooked (since the taking of the rook is not forced) shows only that the puzzle competition was not all that well judged.
you can also go to h6 with the pawn and some more moves and this is so obvious really how no one found it
“For me, it is far better to grasp the Universe as it really is than to persist in delusion, however satisfying and reassuring.” - Carl Sagan
Putting a black pawn on a4 or a white pawn on a4 solves the problem with this puzzle. Stockfish actually finds a bit longer line for black but still loses.
What does rock sacrifice do? Why not move it somewhere else?
I have a question what about rook g4 using that guaranteed check mate?
I tried to read all the comments but there might be a flaw at both stockfish and at first solution. I might be wrong though not entirely sure but if white starts with pawn f7 and continues with rook g4 is there anyway for black to stop check mate? I simply cant see how black could stop it so if any of you have any answers for that please enlighten me
d4 is a discovered check on the white king, and blocks the white bishop. In the long run, the variation is a win for black.
'Thats why the reply cant be black pawn to d4'
Yes, it can, as it is a discovered check against the white king, and gives white to escape through Kg7.
Maybe a black pawn on a4 was missing in the puzzle, which does prevent any possibility of queen attacking the bishop on a1.